Tremendous advances in understanding the molecular alterations contributing to cancer development and progression, and in exploiting this knowledge to build better mouse models that accurately recapitulate many aspects of human cancer have been achieved. New mouse models of human cancer have been used to identify candidate susceptibility genes, targets for cancer therapy and biomarkers for prognosis. However, similarly remarkable advances have not been realized in the clinic, especially for colorectal cancer (CRC), which accounts for the fourth largest number of new cancer cases each year and the second largest number of cancer-related deaths. Rather, large population-based studies of CRC have repeatedly proven that interventions to reduce susceptibility and deployment of early detection programs have the largest impact on survival from CRC. Colorectal cancer is largely preventable with appropriate lifestyle changes and curable if detected early and removed surgically. Building on the knowledge that CRC prevention and early detection are likely to have the greatest impact clinically, we propose a radical new approach to modeling human cancer in mice. We have assembled an experienced team of investigators that will exploit existing mouse models to develop and test innovative approaches for prevention, and robust yet economical methods for early detection of CRC. The foundation of our pioneering approach is a remarkable new mouse population called the Collaborative Cross that accurately models both germline and somatic genetic heterogeneity present within patient populations. We will use this experimentally tractable population-level model with clinically relevant environments to identify robust yet safe approaches for CRC prevention. We will also exploit the cancer heterogeneity of this population-level model, our previous discoveries from large-scale mouse and human CRC comparative gene expression profiling, and the unique ecology of the gastrointestinal tract microbiota to develop passive biosensors for early cancer detection. In parallel, a new biomarker-based mini-cam will be engineered to detect the location of nascent CRCs. RELEVANCE: Our proposed studies are highly relevant to the health of the US population. Colon cancer causes the second largest number of cancer-related deaths. The only proven ways to reduce loss of human life and financial costs of this disease is prevention and early detection. Consequently, innovative new approaches, as presented in this project, are required to reduce the incidence of life-threatening colon cancer.